¿ªÔÆÌåÓý

  1. BITX20
  2. Messages
Expanded Topics Messages Polls
Date
Re: uBITX - A reboot of the old BITX

 

We have to first characterize the crystal. that's measure their motional inductance and capacitance. i do this using the G3URR's method. it is a very simple two transistor oscillator. after that, i use the same oscillator to measure the frequency of each crystal and sort them into bins.
thanks to the bitx40 production, at least for 12 mhz crystals, i don't have to match them. they are kept prematched in separate pouches. i just dip into one!

- f

On Thu, Jun 22, 2017 at 10:00 PM, DJ2AG <ya_duck@...> wrote:
Aashhar, which method are you using to select matching crystals?

73 de Armin, DJ2AG


Re: uBITX - A reboot of the old BITX

 

Aashhar, which method are you using to select matching crystals?

73 de Armin, DJ2AG
Re: uBITX - A reboot of the old BITX

 

"apart from a steady hand" - ah, there's the rub. I was licensed a short while before you, and eyes and hands no longer do what they did. ?But... your advice to practice, practice, practice is excellent, Lawrence! ?I might give some a try. ?Tnx. Charlie K?TAN
Re: uBITX - A reboot of the old BITX

 

hi ashhar

i am following the topics on ubitx with interest

it would be nice if hf sigs did the same as the bitx40 - perhaps more than $59 but other than that a pcb? and set of crystals would be welcome.

if a pcb came about i could simply replace the bitx 40 board and keep the raduino and the enclosure.

updated software as well of cource

best 73 terry gm4dso
Re: uBITX - A reboot of the old BITX

 

Farhan,

i based my analysis on the simple low frequency model of the bipolar transistor as put forth by Hayward in EMRFD and the three relevant white papers. ?I have been working my way through that material, attempting to reproduce Wes' derivations and better understand rf amplifiers. ?This treatment is probably adequate well into the hf region, and perhaps a bit beyond. ?This simple model, which neglects capacitances and other aspects of the BJT, becomes less suitable at higher frequencies of course. ?One great advantage of the smt designs at rf you have put forth is a great reduction in stray reactances and unwanted coupling due to lead lengths and component proximity; very effective grounding is another. ?These issues become quite important at higher frequencies with conventional construction methods using leaded components.

I have not done any circuit modeling, with LT Spice or with the EMRFD software, though I have the first edition of the book with the CD. ?So I shall try that. ?Nor have I actually built the circuits I have discussed! ?I will put together a TIA and see what I can learn. ?I have a bitx40/Raduino here, not yet assembled, while I digest all the software and hardware hacks discussed on this forum and decide how to procede. ?As you have stated, this is truly a Golden Age for homebrew radio! ?Thank you for your contributions. ?I wish I had been a "fly on the wall" at that gathering in Oregon last month.

73,

Ben
Re: Very high clicks after cw wiring

John P
 

On Thu, Jun 22, 2017 at 08:08 am, Allard PE1NWL wrote:
The difference is just in the usage.
Let me try to ask the question differently.

Is it not true that what the BitX board?sees at the PTT connector is the same regardless of whether I press the PTT switch or cause the transistor to conduct via the connection from D7? This is where I'm confused.
?
--
John - WA2FZW
Re: Very high clicks after cw wiring

 

On Thu, June 22, 2017 16:32, John P wrote:
But answer me this if you don't mind. How is pressing the PTT different
than what the transistor does.
The difference is just in the usage.
In the intended usage, in order to return to SSB, the operator would press
the PTT whilst the radio is in RX mode.

But in your case you press the PTT while the radio is still in TX (well it
isn't really but the lockup issue with the RX-TX line keeps it in TX all
the time)

Therefore IMO the first priority would be to fix the lockup issue.

73 Allard PE1NWL
Re: Very high clicks after cw wiring

John P
 

On Thu, Jun 22, 2017 at 05:21 am, Allard PE1NWL wrote:
Did you perhaps switch the emitter and collector of the bypass transistor by mistake? This can lead to similar behaviour (once the transistor is opened it will stay open forever until you power off).
I recommend to look into?this?"lockup issue"?first. I'm quite convinced that once this is fixed the rest will work fine too.

Thanks Allard! That is quite possible.?And it is my plan to do just as you suggested?(just need to find an hour or so to get to it! LOL!).

But answer me this if you don't mind. How is pressing the PTT different than what the transistor does. Sorry, I'm more of a software guy than a hardware guy!

?
?
--
John - WA2FZW
Re: Very high clicks after cw wiring

 

Without the modification when I press the key, the display indicates that it is transmitting in CW mode, but it really isn't
Yes that's correct, since v1.14 you cannot do CW without the TX-RX line

When I press the PTT, which I assume is basically the same as what the transistor in the mod does, the transmit mode on the display switches to LSB
Yes this is as designed:?(since v1.14,??see the revision record in?README.md): CW key down brings the radio?in CW mode automatically. Press the PTT on the mike to return to SSB mode.

I did install the modification but with that in place, when I pressed the key momentarily, the BitX locked up in transmit.
Did you perhaps switch the emitter and collector of the bypass transistor by mistake? This can lead to similar behaviour (once the transistor is opened it will stay open forever until you power off).
I recommend to look into?this?"lockup issue"?first. I'm quite convinced that once this is fixed the rest will work fine too.

73 Allard PE1NWL
Re: Very high clicks after cw wiring

John P
 

I'm having similar issues with CW operation as well. I installed the modification, the modification and the modification. Without the modification when I press the key, the display indicates that it is transmitting in CW mode, but it really isn't. When I press the PTT, which I assume is basically the same as what the transistor in the mod does, the transmit mode on the display switches to LSB and it does transmit, however the carrier level shown on the scope is far less than I would expect.

I did install the modification but with that in place, when I pressed the key momentarily, the BitX locked up in transmit (don't remember which mode). I t could have been a bad transistor, or some other issue. After that happened, I removed the modification. I still need to play with it more before I scream for real help and I want to take a look to see how the software handles CW mode.



--
John - WA2FZW
Re: Any info about making air variable capacitors, and magnetic loops?

Vince Vielhaber
 

If you're on Facebook, there are two groups dedicated to magnetic loop
antennas.

Vince.

Thank you everyone. I got a lot of new info from here and google. Nobody
around here has any experience with the mag loop, but there are two of us,
myself and an extra class operator who are planning a build for the
experience. And he is the person who recommended the BITX40 to me in the
first place. But he hasn't assembled either of his two analog and digital
BITXes. So, I am not to concerned about him constructing one first. Hi Hi.
Even a little friendly competition can help improve one's amateur radio
skills. Thanks again.
--
Michigan VHF Corp.

Re: uBITX - A reboot of the old BITX

 

excellent idea Ashar. thanks

73

ken VA3ABN

On Thu, Jun 22, 2017 at 5:27 AM, Ashhar Farhan <farhanbox@...> wrote:
with the ubitx, i have extensively used the EMRFD tools : the ladbuild, the gpla, the fba and the cascade. maybe i should do a video that explains how to use it.

- f

On Thu, Jun 22, 2017 at 1:33 PM, Trent Trent <vk7hrs@...> wrote:
Is that how you designed the bitx

On 22 Jun. 2017 3:13 pm, "Ashhar Farhan" <farhanbox@...> wrote:
Here is a screenshot of my simulation. i run it under ubuntu, hence the weird window style.

- f

On Thu, Jun 22, 2017 at 10:06 AM, Ashhar Farhan <farhanbox@...> wrote:
ben,
i simulated the amps using fba.exe that accompanies the EMRFD cd. you can download this and other goodies from . it is somewhere under the EMRFD link from the home page.?
i do remember that the match was very close to 50 input. but you could be right. the more complicating problem is that the operating frequency for this amp is vhf. the simple relationship of the resistor ratios doesnt hold up there.
i will build and measure the return loss at 50 ohms soon and let you know. thanks for pointing this out.

- f

On 22 Jun 2017 7:40 a.m., "Ben Bangerter, K0IKR via Groups.Io" <bwbangerter=yahoo.com@groups.io> wrote:

The design that Farhan has put forth for the ubitx transceiver project utilizes bi-directional termination-insensitive amplifiers (TIAs) based on work done by Wes Hayward, w7zoi, and Bob Kopski, k7nhi, and described in the white paper ¡°A Termination Insensitive Amplifier for Bidirectional Transceivers,¡± which can be found on Wes¡¯ web site: .? Feedback amplifiers of this type have stable and predictable gains, and most importantly, well-defined input and output impedances, which are most important for terminating filters and mixers.

I have been studying the ubitx schematic in some detail, and I conclude that the TIA shown has an input impedance of ~100 ohms.? Lets look at RXamp1, which employs Q10, Q11 and Q12, and is paired with what we may call TXamp2, with Q20, Q21, and Q22.? The input impedance as calculated by Farhan¡¯s approximate relation is:

?Zin = ?R11xR13/R14 = 100 ohms,

And the impedance as calculated by Hayward¡¯s more accurate relation (described in his white paper ¡°The Feedback Amplifier with a Simple Model,¡± available on his web site) is:

?Zin = R13x(R14+R11)/(R14+R13) = 105 ohms.

When RXamp1 is paired with TXamp2 to create the bi-directional construct, in receive mode (when RXamp1 is powered and TXamp2 is not) the input to RXamp1 is in parallel with a load (for RF) of the series combination of R26+R27 = 517 ohms.? As a consequence, the output of mixer 1 is terminated in an impedance of 84 ohms (100||517) rather than the desired 50 ohms.? Actually, there is additional loading by R105 in the keying circuit, which lowers the impedance seen by the mixer to 77 ohms.? The output impedance of the bi-directional TIA in receive mode is 50 ohms in parallel with 708 ohms from the combination of R20, R21, and R24, or 47 ohms, which is the impedance presented to the input of the 45 MHz bandpass filter.

In transmit mode, my analysis shows that the 45 MHz bandpass filter is terminated in 84 ohms, and the impedance presented to the first mixer is 45 ohms.

The bi-directional TIA formed by RXamp2 and TXamp1 has the same departure from ideal behavior.? In receive mode, the second mixer is terminated in 84 ohms, and in transmit mode the 12 MHz crystal filter is terminated in 84 ohms.

Does all this have any significant effect on how well the ubitx performs?? Perhaps not.? I think the effect of non-ideal termination of an LC or crystal filter affects mainly passband ripple, and non-ideal termination of a mixer affects the amplitudes of mixing products and may increase the amplitudes of ¡°birdies¡± in the receiver.

If I got ¡°lost in the weeds¡± with my analysis, please set me straight!
Re: uBITX - A reboot of the old BITX

 

The ubitx is released. go and make it!

- f

On Thu, Jun 22, 2017 at 3:07 PM, Trent Trent <vk7hrs@...> wrote:
A vid would be nice and there would be a fsir bit of work to use it all,btw when is the release date for the radio

On 22 Jun. 2017 7:27 pm, "Ashhar Farhan" <farhanbox@...> wrote:
with the ubitx, i have extensively used the EMRFD tools : the ladbuild, the gpla, the fba and the cascade. maybe i should do a video that explains how to use it.

- f

On Thu, Jun 22, 2017 at 1:33 PM, Trent Trent <vk7hrs@...> wrote:
Is that how you designed the bitx

On 22 Jun. 2017 3:13 pm, "Ashhar Farhan" <farhanbox@...> wrote:
Here is a screenshot of my simulation. i run it under ubuntu, hence the weird window style.

- f

On Thu, Jun 22, 2017 at 10:06 AM, Ashhar Farhan <farhanbox@...> wrote:
ben,
i simulated the amps using fba.exe that accompanies the EMRFD cd. you can download this and other goodies from . it is somewhere under the EMRFD link from the home page.?
i do remember that the match was very close to 50 input. but you could be right. the more complicating problem is that the operating frequency for this amp is vhf. the simple relationship of the resistor ratios doesnt hold up there.
i will build and measure the return loss at 50 ohms soon and let you know. thanks for pointing this out.

- f

On 22 Jun 2017 7:40 a.m., "Ben Bangerter, K0IKR via Groups.Io" <bwbangerter=yahoo.com@groups.io> wrote:

The design that Farhan has put forth for the ubitx transceiver project utilizes bi-directional termination-insensitive amplifiers (TIAs) based on work done by Wes Hayward, w7zoi, and Bob Kopski, k7nhi, and described in the white paper ¡°A Termination Insensitive Amplifier for Bidirectional Transceivers,¡± which can be found on Wes¡¯ web site: .? Feedback amplifiers of this type have stable and predictable gains, and most importantly, well-defined input and output impedances, which are most important for terminating filters and mixers.

I have been studying the ubitx schematic in some detail, and I conclude that the TIA shown has an input impedance of ~100 ohms.? Lets look at RXamp1, which employs Q10, Q11 and Q12, and is paired with what we may call TXamp2, with Q20, Q21, and Q22.? The input impedance as calculated by Farhan¡¯s approximate relation is:

?Zin = ?R11xR13/R14 = 100 ohms,

And the impedance as calculated by Hayward¡¯s more accurate relation (described in his white paper ¡°The Feedback Amplifier with a Simple Model,¡± available on his web site) is:

?Zin = R13x(R14+R11)/(R14+R13) = 105 ohms.

When RXamp1 is paired with TXamp2 to create the bi-directional construct, in receive mode (when RXamp1 is powered and TXamp2 is not) the input to RXamp1 is in parallel with a load (for RF) of the series combination of R26+R27 = 517 ohms.? As a consequence, the output of mixer 1 is terminated in an impedance of 84 ohms (100||517) rather than the desired 50 ohms.? Actually, there is additional loading by R105 in the keying circuit, which lowers the impedance seen by the mixer to 77 ohms.? The output impedance of the bi-directional TIA in receive mode is 50 ohms in parallel with 708 ohms from the combination of R20, R21, and R24, or 47 ohms, which is the impedance presented to the input of the 45 MHz bandpass filter.

In transmit mode, my analysis shows that the 45 MHz bandpass filter is terminated in 84 ohms, and the impedance presented to the first mixer is 45 ohms.

The bi-directional TIA formed by RXamp2 and TXamp1 has the same departure from ideal behavior.? In receive mode, the second mixer is terminated in 84 ohms, and in transmit mode the 12 MHz crystal filter is terminated in 84 ohms.

Does all this have any significant effect on how well the ubitx performs?? Perhaps not.? I think the effect of non-ideal termination of an LC or crystal filter affects mainly passband ripple, and non-ideal termination of a mixer affects the amplitudes of mixing products and may increase the amplitudes of ¡°birdies¡± in the receiver.

If I got ¡°lost in the weeds¡± with my analysis, please set me straight!
Re: uBitx Prototype (PCB)

 

hi
what size is the board

terry gm4dso
Re: uBITX - A reboot of the old BITX

Trent Trent
 

A vid would be nice and there would be a fsir bit of work to use it all,btw when is the release date for the radio

On 22 Jun. 2017 7:27 pm, "Ashhar Farhan" <farhanbox@...> wrote:
with the ubitx, i have extensively used the EMRFD tools : the ladbuild, the gpla, the fba and the cascade. maybe i should do a video that explains how to use it.

- f

On Thu, Jun 22, 2017 at 1:33 PM, Trent Trent <vk7hrs@...> wrote:
Is that how you designed the bitx

On 22 Jun. 2017 3:13 pm, "Ashhar Farhan" <farhanbox@...> wrote:
Here is a screenshot of my simulation. i run it under ubuntu, hence the weird window style.

- f

On Thu, Jun 22, 2017 at 10:06 AM, Ashhar Farhan <farhanbox@...> wrote:
ben,
i simulated the amps using fba.exe that accompanies the EMRFD cd. you can download this and other goodies from . it is somewhere under the EMRFD link from the home page.?
i do remember that the match was very close to 50 input. but you could be right. the more complicating problem is that the operating frequency for this amp is vhf. the simple relationship of the resistor ratios doesnt hold up there.
i will build and measure the return loss at 50 ohms soon and let you know. thanks for pointing this out.

- f

On 22 Jun 2017 7:40 a.m., "Ben Bangerter, K0IKR via Groups.Io" <bwbangerter=yahoo.com@groups.io> wrote:

The design that Farhan has put forth for the ubitx transceiver project utilizes bi-directional termination-insensitive amplifiers (TIAs) based on work done by Wes Hayward, w7zoi, and Bob Kopski, k7nhi, and described in the white paper ¡°A Termination Insensitive Amplifier for Bidirectional Transceivers,¡± which can be found on Wes¡¯ web site: .? Feedback amplifiers of this type have stable and predictable gains, and most importantly, well-defined input and output impedances, which are most important for terminating filters and mixers.

I have been studying the ubitx schematic in some detail, and I conclude that the TIA shown has an input impedance of ~100 ohms.? Lets look at RXamp1, which employs Q10, Q11 and Q12, and is paired with what we may call TXamp2, with Q20, Q21, and Q22.? The input impedance as calculated by Farhan¡¯s approximate relation is:

?Zin = ?R11xR13/R14 = 100 ohms,

And the impedance as calculated by Hayward¡¯s more accurate relation (described in his white paper ¡°The Feedback Amplifier with a Simple Model,¡± available on his web site) is:

?Zin = R13x(R14+R11)/(R14+R13) = 105 ohms.

When RXamp1 is paired with TXamp2 to create the bi-directional construct, in receive mode (when RXamp1 is powered and TXamp2 is not) the input to RXamp1 is in parallel with a load (for RF) of the series combination of R26+R27 = 517 ohms.? As a consequence, the output of mixer 1 is terminated in an impedance of 84 ohms (100||517) rather than the desired 50 ohms.? Actually, there is additional loading by R105 in the keying circuit, which lowers the impedance seen by the mixer to 77 ohms.? The output impedance of the bi-directional TIA in receive mode is 50 ohms in parallel with 708 ohms from the combination of R20, R21, and R24, or 47 ohms, which is the impedance presented to the input of the 45 MHz bandpass filter.

In transmit mode, my analysis shows that the 45 MHz bandpass filter is terminated in 84 ohms, and the impedance presented to the first mixer is 45 ohms.

The bi-directional TIA formed by RXamp2 and TXamp1 has the same departure from ideal behavior.? In receive mode, the second mixer is terminated in 84 ohms, and in transmit mode the 12 MHz crystal filter is terminated in 84 ohms.

Does all this have any significant effect on how well the ubitx performs?? Perhaps not.? I think the effect of non-ideal termination of an LC or crystal filter affects mainly passband ripple, and non-ideal termination of a mixer affects the amplitudes of mixing products and may increase the amplitudes of ¡°birdies¡± in the receiver.

If I got ¡°lost in the weeds¡± with my analysis, please set me straight!
Re: uBITX - A reboot of the old BITX

 

with the ubitx, i have extensively used the EMRFD tools : the ladbuild, the gpla, the fba and the cascade. maybe i should do a video that explains how to use it.

- f

On Thu, Jun 22, 2017 at 1:33 PM, Trent Trent <vk7hrs@...> wrote:
Is that how you designed the bitx

On 22 Jun. 2017 3:13 pm, "Ashhar Farhan" <farhanbox@...> wrote:
Here is a screenshot of my simulation. i run it under ubuntu, hence the weird window style.

- f

On Thu, Jun 22, 2017 at 10:06 AM, Ashhar Farhan <farhanbox@...> wrote:
ben,
i simulated the amps using fba.exe that accompanies the EMRFD cd. you can download this and other goodies from . it is somewhere under the EMRFD link from the home page.?
i do remember that the match was very close to 50 input. but you could be right. the more complicating problem is that the operating frequency for this amp is vhf. the simple relationship of the resistor ratios doesnt hold up there.
i will build and measure the return loss at 50 ohms soon and let you know. thanks for pointing this out.

- f

On 22 Jun 2017 7:40 a.m., "Ben Bangerter, K0IKR via Groups.Io" <bwbangerter=yahoo.com@groups.io> wrote:

The design that Farhan has put forth for the ubitx transceiver project utilizes bi-directional termination-insensitive amplifiers (TIAs) based on work done by Wes Hayward, w7zoi, and Bob Kopski, k7nhi, and described in the white paper ¡°A Termination Insensitive Amplifier for Bidirectional Transceivers,¡± which can be found on Wes¡¯ web site: .? Feedback amplifiers of this type have stable and predictable gains, and most importantly, well-defined input and output impedances, which are most important for terminating filters and mixers.

I have been studying the ubitx schematic in some detail, and I conclude that the TIA shown has an input impedance of ~100 ohms.? Lets look at RXamp1, which employs Q10, Q11 and Q12, and is paired with what we may call TXamp2, with Q20, Q21, and Q22.? The input impedance as calculated by Farhan¡¯s approximate relation is:

?Zin = ?R11xR13/R14 = 100 ohms,

And the impedance as calculated by Hayward¡¯s more accurate relation (described in his white paper ¡°The Feedback Amplifier with a Simple Model,¡± available on his web site) is:

?Zin = R13x(R14+R11)/(R14+R13) = 105 ohms.

When RXamp1 is paired with TXamp2 to create the bi-directional construct, in receive mode (when RXamp1 is powered and TXamp2 is not) the input to RXamp1 is in parallel with a load (for RF) of the series combination of R26+R27 = 517 ohms.? As a consequence, the output of mixer 1 is terminated in an impedance of 84 ohms (100||517) rather than the desired 50 ohms.? Actually, there is additional loading by R105 in the keying circuit, which lowers the impedance seen by the mixer to 77 ohms.? The output impedance of the bi-directional TIA in receive mode is 50 ohms in parallel with 708 ohms from the combination of R20, R21, and R24, or 47 ohms, which is the impedance presented to the input of the 45 MHz bandpass filter.

In transmit mode, my analysis shows that the 45 MHz bandpass filter is terminated in 84 ohms, and the impedance presented to the first mixer is 45 ohms.

The bi-directional TIA formed by RXamp2 and TXamp1 has the same departure from ideal behavior.? In receive mode, the second mixer is terminated in 84 ohms, and in transmit mode the 12 MHz crystal filter is terminated in 84 ohms.

Does all this have any significant effect on how well the ubitx performs?? Perhaps not.? I think the effect of non-ideal termination of an LC or crystal filter affects mainly passband ripple, and non-ideal termination of a mixer affects the amplitudes of mixing products and may increase the amplitudes of ¡°birdies¡± in the receiver.

If I got ¡°lost in the weeds¡± with my analysis, please set me straight!
Re: uBitx Prototype (PCB)

 

hi

is your pcb all thru hole . what are the transformers , somebody has suggested 10 turns on ft37-43 is that right

terry gm4dso
Re: uBITX - A reboot of the old BITX

Trent Trent
 

Is that how you designed the bitx

On 22 Jun. 2017 3:13 pm, "Ashhar Farhan" <farhanbox@...> wrote:
Here is a screenshot of my simulation. i run it under ubuntu, hence the weird window style.

- f

On Thu, Jun 22, 2017 at 10:06 AM, Ashhar Farhan <farhanbox@...> wrote:
ben,
i simulated the amps using fba.exe that accompanies the EMRFD cd. you can download this and other goodies from . it is somewhere under the EMRFD link from the home page.?
i do remember that the match was very close to 50 input. but you could be right. the more complicating problem is that the operating frequency for this amp is vhf. the simple relationship of the resistor ratios doesnt hold up there.
i will build and measure the return loss at 50 ohms soon and let you know. thanks for pointing this out.

- f

On 22 Jun 2017 7:40 a.m., "Ben Bangerter, K0IKR via Groups.Io" <bwbangerter=yahoo.com@groups.io> wrote:

The design that Farhan has put forth for the ubitx transceiver project utilizes bi-directional termination-insensitive amplifiers (TIAs) based on work done by Wes Hayward, w7zoi, and Bob Kopski, k7nhi, and described in the white paper ¡°A Termination Insensitive Amplifier for Bidirectional Transceivers,¡± which can be found on Wes¡¯ web site: .? Feedback amplifiers of this type have stable and predictable gains, and most importantly, well-defined input and output impedances, which are most important for terminating filters and mixers.

I have been studying the ubitx schematic in some detail, and I conclude that the TIA shown has an input impedance of ~100 ohms.? Lets look at RXamp1, which employs Q10, Q11 and Q12, and is paired with what we may call TXamp2, with Q20, Q21, and Q22.? The input impedance as calculated by Farhan¡¯s approximate relation is:

?Zin = ?R11xR13/R14 = 100 ohms,

And the impedance as calculated by Hayward¡¯s more accurate relation (described in his white paper ¡°The Feedback Amplifier with a Simple Model,¡± available on his web site) is:

?Zin = R13x(R14+R11)/(R14+R13) = 105 ohms.

When RXamp1 is paired with TXamp2 to create the bi-directional construct, in receive mode (when RXamp1 is powered and TXamp2 is not) the input to RXamp1 is in parallel with a load (for RF) of the series combination of R26+R27 = 517 ohms.? As a consequence, the output of mixer 1 is terminated in an impedance of 84 ohms (100||517) rather than the desired 50 ohms.? Actually, there is additional loading by R105 in the keying circuit, which lowers the impedance seen by the mixer to 77 ohms.? The output impedance of the bi-directional TIA in receive mode is 50 ohms in parallel with 708 ohms from the combination of R20, R21, and R24, or 47 ohms, which is the impedance presented to the input of the 45 MHz bandpass filter.

In transmit mode, my analysis shows that the 45 MHz bandpass filter is terminated in 84 ohms, and the impedance presented to the first mixer is 45 ohms.

The bi-directional TIA formed by RXamp2 and TXamp1 has the same departure from ideal behavior.? In receive mode, the second mixer is terminated in 84 ohms, and in transmit mode the 12 MHz crystal filter is terminated in 84 ohms.

Does all this have any significant effect on how well the ubitx performs?? Perhaps not.? I think the effect of non-ideal termination of an LC or crystal filter affects mainly passband ripple, and non-ideal termination of a mixer affects the amplitudes of mixing products and may increase the amplitudes of ¡°birdies¡± in the receiver.

If I got ¡°lost in the weeds¡± with my analysis, please set me straight!
Re: Bitx connector pins

EA3IAV
 

Is there anywhere on the internet where i can get them already mounted? I need several sets of 2 connector wires to tidy up the wiring thing
thanks
Re: uBITX - A reboot of the old BITX

 

Here is a screenshot of my simulation. i run it under ubuntu, hence the weird window style.

- f

On Thu, Jun 22, 2017 at 10:06 AM, Ashhar Farhan <farhanbox@...> wrote:
ben,
i simulated the amps using fba.exe that accompanies the EMRFD cd. you can download this and other goodies from . it is somewhere under the EMRFD link from the home page.?
i do remember that the match was very close to 50 input. but you could be right. the more complicating problem is that the operating frequency for this amp is vhf. the simple relationship of the resistor ratios doesnt hold up there.
i will build and measure the return loss at 50 ohms soon and let you know. thanks for pointing this out.

- f

On 22 Jun 2017 7:40 a.m., "Ben Bangerter, K0IKR via Groups.Io" <bwbangerter=yahoo.com@groups.io> wrote:

The design that Farhan has put forth for the ubitx transceiver project utilizes bi-directional termination-insensitive amplifiers (TIAs) based on work done by Wes Hayward, w7zoi, and Bob Kopski, k7nhi, and described in the white paper ¡°A Termination Insensitive Amplifier for Bidirectional Transceivers,¡± which can be found on Wes¡¯ web site: .? Feedback amplifiers of this type have stable and predictable gains, and most importantly, well-defined input and output impedances, which are most important for terminating filters and mixers.

I have been studying the ubitx schematic in some detail, and I conclude that the TIA shown has an input impedance of ~100 ohms.? Lets look at RXamp1, which employs Q10, Q11 and Q12, and is paired with what we may call TXamp2, with Q20, Q21, and Q22.? The input impedance as calculated by Farhan¡¯s approximate relation is:

?Zin = ?R11xR13/R14 = 100 ohms,

And the impedance as calculated by Hayward¡¯s more accurate relation (described in his white paper ¡°The Feedback Amplifier with a Simple Model,¡± available on his web site) is:

?Zin = R13x(R14+R11)/(R14+R13) = 105 ohms.

When RXamp1 is paired with TXamp2 to create the bi-directional construct, in receive mode (when RXamp1 is powered and TXamp2 is not) the input to RXamp1 is in parallel with a load (for RF) of the series combination of R26+R27 = 517 ohms.? As a consequence, the output of mixer 1 is terminated in an impedance of 84 ohms (100||517) rather than the desired 50 ohms.? Actually, there is additional loading by R105 in the keying circuit, which lowers the impedance seen by the mixer to 77 ohms.? The output impedance of the bi-directional TIA in receive mode is 50 ohms in parallel with 708 ohms from the combination of R20, R21, and R24, or 47 ohms, which is the impedance presented to the input of the 45 MHz bandpass filter.

In transmit mode, my analysis shows that the 45 MHz bandpass filter is terminated in 84 ohms, and the impedance presented to the first mixer is 45 ohms.

The bi-directional TIA formed by RXamp2 and TXamp1 has the same departure from ideal behavior.? In receive mode, the second mixer is terminated in 84 ohms, and in transmit mode the 12 MHz crystal filter is terminated in 84 ohms.

Does all this have any significant effect on how well the ubitx performs?? Perhaps not.? I think the effect of non-ideal termination of an LC or crystal filter affects mainly passband ripple, and non-ideal termination of a mixer affects the amplitudes of mixing products and may increase the amplitudes of ¡°birdies¡± in the receiver.

If I got ¡°lost in the weeds¡± with my analysis, please set me straight!



ubtix_fba_simulation.png

88261 - 88280 of 116949